Penerapan Convolutional Neural Network Untuk Klasifikasi Kanker Kulit Melanoma Pada Dataset Gambar Kulit
Keywords:
media sosial, food vlogger, elemen citra, pandemi covid-19Abstract
Melanoma skin cancer is one of the most dangerous skin cancers where the ferocity and speed of metastasis has caused a high mortality rate among afflicted when the cancer is not treated. Early detection of the cancer and prevention by removing the affected skin have been shown to decrease the mortality rate on afflicted patient. Thus, development of a method to help automatically diagnose the cancer and classify between cancer and normal mole or birthmark is needed. Previous methods still show limitations in classifying melanoma skin cancer. This study proposes a classification system using convolutional neural network trained on the original ISIC 2020 dataset and hair removed dataset which is then combined using ensemble.
The dataset used is first preprocessed using the hair removal algorithm convolutional neural network using EfficientNet B0 – B7 and ResNet-50-v2 will be trained using ISIC 2020 dataset and ISIC 2020 dataset processed with hair removal algorithm.The model is evaluated using test data from ISIC 2020 dataset on area under the receiver operating characteristic curve (ROC AUC). The model trained will then be combined using ensemble where the result of the model will be averaged to give a combined prediction.
The result of the test shows that the model trained is capable to classify melanoma and non-melanoma images. It is also shown that by removing hair from the skin image reduces the accuracy of th e model. Using Ensembling on the different models trained into one meta-model also increases the accuracy of the prediction giving a high final accuracy of 93.108%.
References
[1] Al-masni, M.A., Kim, D.H. and Kim, T.S. 2020. Multiple
skin lesions diagnostics via integrated deep convolutional
networks for segmentation and classification. Computer
Methods and Programs in Biomedicine. 190.
DOI:https://doi.org/10.1016/j.cmpb.2020.105351.
[2] Bray, F., Ferlay, J., Soerjomataram, I., Siegel, R.L., Torre,
L.A. and Jemal, A. 2018. Global cancer statistics 2018:
GLOBOCAN estimates of incidence and mortality
worldwide for 36 cancers in 185 countries. CA: A Cancer
Journal for Clinicians. 68, 6, 394–424.
DOI:https://doi.org/10.3322/caac.21492.
[3] Domingos, P. 1997. Why Does Bagging Work? A Bayesian
Account and its Implications. Proceedings of the Third
International Conference on Knowledge Discovery and
Data Mining (KDD’97) (Newport Beach, CA), 158.
[4] El-Khatib, H., Popescu, D. and Ichim, L. 2020. Deep
learning–based methods for automatic diagnosis of skin
lesions. Sensors (Switzerland). 20, 6.
DOI:https://doi.org/10.3390/s20061753.
[5] Gessert, N., Nielsen, M., Shaikh, M., Werner, R. and
Schlaefer, A. 2020. Skin lesion classification using
ensembles of multi-resolution EfficientNets with meta data.
MethodsX. 7, 100864.
DOI:https://doi.org/10.1016/j.mex.2020.100864.
[6] Grandvalet, Y. 2004. Bagging equalizes influence. Machine
Learning. 55, 3, 251–270.
DOI:https://doi.org/10.1023/B:MACH.0000027783.34431.4
2.
[7] Guo, P., Xue, Z., Mtema, Z., Yeates, K., Ginsburg, O.,
Demarco, M., Rodney Long, L., Schiffman, M. and Antani,
S. 2020. Ensemble deep learning for cervix image selection
toward improving reliability in automated cervical
precancer screening. Diagnostics. 10, 7.
DOI:https://doi.org/10.3390/diagnostics10070451.
[8] He, K., Zhang, X., Ren, S. and Sun, J. 2016. Identity
mappings in deep residual networks. Lecture Notes in
Computer Science (including subseries Lecture Notes in
Artificial Intelligence and Lecture Notes in Bioinformatics),
630–645.
[9] Heisler, M., Karst, S., Lo, J., Mammo, Z., Yu, T., Warner,
S., Maberley, D., Beg, M.F., Navajas, E. V. and Sarunic, M.
V. 2020. Ensemble deep learning for diabetic retinopathy
detection using optical coherence tomography angiography.
Translational Vision Science and Technology. 9, 2, 1–11.
DOI:https://doi.org/10.1167/tvst.9.2.20.
[10] Hosny, K.M., Kassem, M.A. and Fouad, M.M. 2020.
Classification of Skin Lesions into Seven Classes Using
Transfer Learning with AlexNet. Journal of Digital
Imaging.1–10. DOI:https://doi.org/10.1007/s10278-020-
00371-9.
[11] Jerant, A.F., Johnson, J.T., Sheridan, C.D. and Caffrey, T.J.
2000. Early Detection and Treatment of Skin Cancer.
American Family Physician. 62, 2, 357–368.
[12] Koehoorn, J., Sobiecki, A., Rauber, P., Jalba, A. and Telea,
A. 2016. Effcient and Effective Automated Digital Hair
Removal from Dermoscopy Images. Mathematical
Morphology - Theory and Applications. 1, 1, 1–17.
DOI:https://doi.org/10.1515/mathm-2016-0001.
[13] Neto, H.A., Tavares, W.L.F., Ribeiro, D.C.S.Z., Alves,
R.C.O., Fonseca, L.M. and Campos, S.V.A. 2019. On the
utilization of deep and ensemble learning to detect milk
adulteration. BioData Mining. 12, 1.
DOI:https://doi.org/10.1186/s13040-019-0200-5.
[14] Premaladha, J. and Ravichandran, K.S. 2016. Novel
Approaches for Diagnosing Melanoma Skin Lesions
Through Supervised and Deep Learning Algorithms.
Journal of Medical Systems. 40, 4, 1–12.
DOI:https://doi.org/10.1007/s10916-016-0460-2.
[15] Rotemberg, V. et al. 2020. A Patient-Centric Dataset of
Images and Metadata for Identifying Melanomas Using
Clinical Context.
[16] Siegel, R.L., Miller, K.D. and Jemal, A. 2020. Cancer
statistics, 2020. CA: A Cancer Journal for Clinicians. 70, 1,
7–30. DOI:https://doi.org/10.3322/caac.21590.
[17] Tan, M. and Le, Q. V. 2019. EfficientNet: Rethinking model
scaling for convolutional neural networks. 36th
International Conference on Machine Learning, ICML
2019, 10691–10700.
[18] Wei, X., Gao, M., Yu, R., Liu, Z., Gu, Q., Liu, X., Zheng,
Z., Zheng, X., Zhu, J. and Zhang, S. 2020. Ensemble deep
learning model for multicenter classification of thyroid
nodules on ultrasound images. Medical Science Monitor. 26,
. DOI:https://doi.org/10.12659/MSM.926096.
[19] Yosinski, J., Clune, J., Bengio, Y. and Lipson, H. 2014.
How transferable are features in deep neural networks?
Advances in Neural Information Processing Systems. 4,
January, 3320–3328.
[20] Zunair, H. and Ben Hamza, A. 2020. Melanoma detection
using adversarial training and deep transfer learning.
Physics in Medicine and Biology. 65, 13, 135005.
